Pinned Loading

1. debate-degrades-reasoning debate-degrades-reasoning Public

   Single-round debate degrades LLM reasoning in symmetric settings — 2,100 evaluations, 11 conditions, two benchmarks

   Python

2. receipt-gated-pipelines receipt-gated-pipelines Public

   Receipt-Gated Pipelines: Cryptographic Verification of Tool-Call Claims in Multi-Agent LLM Systems

   Python

3. structure-beats-scale structure-beats-scale Public

   Structure Beats Scale: How Structured Review Outperforms Brute-Force Generation in LLM Code Synthesis

   Python